Residues of organophosphate insecticides are undesirable contaminants of the environment and a range of commodities. Areas of particular sensitivity include contamination of soil, irrigation tailwater that is re-cycled, used by irrigators downstream or simply allowed to run off-farm, and residues above permissible levels in agricultural and horticultural exports. Poisoning with organophosphates presents a problem for agricultural workers that are exposed to these chemicals, as well as military personnel exposed to organophosphates used in chemical warfare. Furthermore, the stockpiling of organophosphorus nerve agents has resulted in the need to detoxify these stocks. Bioremediation strategies are therefore required for eliminating or reducing these organophosphate residues and/or stockpiles.
One proposed strategy involves the use of enzymes capable of immobilising or degrading the organophosphate residues. Such enzymes may be employed, for example, in bioreactors through which contaminated water could be passed, or in washing solutions after post-harvest disinfestation of fruit, vegetables or animal products to reduce residue levels and withholding times. Suitable enzymes for degrading organophosphate residues include OP hydrolases from bacteria (Mulbry, 1992; Mulbry and Kearney, 1991; Cheng et al., 1999; U.S. Pat. No. 5,484,728; U.S. Pat. No. 5,589,386), vertebrates (Wang et al., 1993; 1998; Gan et al, 1991; Broomfield et al., 1999) and OP resistant insects (WO 95/19440 and WO 97/19176). It is desirable that the OP hydrolases degrade the organophosphate residues at a rapid rate.
The most thoroughly studied OP degrading enzyme is bacterial organophosphate dihydrolase (OPD), which is encoded by identical genes on dissimilar plasmids in both Flavobacterium sp. ATCC 27551 and Brevundimonas diminuta MG (Harper et al., 1988; Mulbry and Karns, 1989). OPD is a homodimeric protein that is capable of hydrolysing a wide range of phosphate triesters (both oxon and thion OPs) (Dumas et al., 1989a, b). Its reaction mechanism directly or indirectly involves metal ions, preferably Zn++. OPD has no detectable activity with phosphate monoesters or diesters (Dumas et al., 1989a, b; 1990).
OPD homologues (phosphotriesterase homology proteins, or PHPs) have been identified in the genomes of Escherichia coli (ePHP), Mycobacterium tuberculosis (mtPHP) and Mycoplasma pneumoniae (mpPHP), although only ePHP has been tested for phosphotriesterase activity (Scanlan and Reid, 1995; Buchbinder et al., 1998). No activity was detected in ePHP crude lysates with any of the substrates tested, such as p-nitrophenyl acetate, bis(p-nitrophenyl) phosphate, paraoxon and p-nitrophenyl phosphate.
OPD homologues have also been identified in vertebrates (Davies et al., 1997), although their function in these organisms is unknown. OPD, ePHP, mtPHP and mammalian PHPs are 27-30% identical at the amino acid level, while mpPHP is less similar. Amino acid residues involved in Zn++ binding are conserved across the six members of the phosphotriesterase family identified to date (Buchbinder et al., 1998).
Three other distinct OP hydrolysing enzymes have been isolated from bacteria with a history of exposure to OPs (Mulbry and Karns, 1989; Mulbry, 1992; Cheng et al., 1999). The two for which sequence data are available are unrelated to each other and to OPD. One, a prolidase from Alteromonas sp., normally functions in hydrolysis of X-Pro dipeptides. Its activity for insecticidal OPs is reported as modest, although it has not been reported in terms of kcat/Km specificity constants (Cheng et al., 1999). The other, an aryldialkylphosphatase (ADPase) from Nocardia sp. strain B-1, has a turnover number for ethyl parathion that is 4500-fold lower than that reported for OPD (Mulbry and Karns, 1989; Mulbry, 1992).
Paraoxonase, or PON1, is a distinct OP hydrolysing enzyme found in mammals. Like OPD it is a metalloenzyme, preferring Ca++ in this case, which is associated with low density lipoproteins in plasma and normally involved in metabolism of oxidised lipid compounds (Gan et al., 1991; Sorenson et al., 1995). It has high activity for paraoxon, with a specificity constant of around 106M−1 sec−1 (Doom et al., 1999; Hong and Raushel, 1999).
There is also evidence for other, so-called diisopropyl fluorophosphatase (DFPase) enzymes in a wide range of vertebrates, invertebrates and microorganisms (Wang et al., 1998; Hoskin et al., 1999; Billecke et al., 1999). These enzymes are notably diverse in many of their biochemical properties but are all characterised by their hydrolytic activity against OP chemical warfare agents. Limited sequence data suggest that they are unrelated to all the other OP hydrolytic enzymes described above.
OP resistant blowflies and houseflies have been the source of esterase enzymes with activity against oxon OPs like chlorfenvinfos (CVP) and carboxylester OPs like malathion (Newcomb et al., 1997; Campbell et al. 1998; Claudianos et al. 1999; WO 95/19440; WO 97/19176). A Gly to Asp substitution at residue 137 in blowfly esterase E3 (and its housefly ortholog, ALI) resulted in the acquisition of activity for CVP, while a Trp to Leu/Ser mutation at residue 251 in the same enzyme resulted in activity against malathion. However, the specificity constants of these enzymes for their OP substrates are orders of magnitude less than those of OPD for paraoxon.
There is a need for further OP degrading enzymes which can be used in bioremediation strategies.